Minimum Wage and Teen Labor Participation

ISEF Category: Behavioral and Social Sciences

Subcategory: Sociology and Anthropology  ·  Difficulty: Advanced  ·  Setup: University Lab  ·  Time: Full Year

The Hook

When a state raises the minimum wage, teen jobs can shift fast. A pay rule can change who works, who keeps looking, and who leaves the labor market. You can study that ripple with public survey data instead of running a live policy test. That makes a real policy question feel like a clean research project.

What Is It?

Labor-force participation means the share of teens who are working or looking for work. The Current Population Survey, or CPS, is a large U.S. household survey that records work status, age, school status, and other details. With CPS microdata, you can track teen labor patterns before and after a state wage change.

Synthetic control is a way to build a look-alike version of one state from a weighted mix of other states. Think of it like making a comparison twin from several neighbors. If the treated state starts to move away from its synthetic twin after the policy date, that gap gives you an estimate of the policy effect.

Why This Is a Good Topic

This is a strong science fair topic because you can test a real policy change with public data, clear outcomes, and reproducible math. It connects to jobs, school, and family finances, so the result matters outside class. You can learn data cleaning, causal inference, and how to defend a comparison group, all with tools a motivated student can handle.

Research Questions

• How does a state minimum-wage increase change the labor-force participation rate of 16- to 19-year-olds in the treated state?
• What is the effect of a minimum-wage increase on teen labor-force participation compared with a matched synthetic control state?
• Does the policy effect differ between 16- and 17-year-olds and 18- to 19-year-olds?
• To what extent do changes vary by school enrollment status or part-time work status?
• Which donor states produce the closest pre-policy fit for the synthetic control?
• How does the estimated effect change when you swap the donor pool or run placebo states?

Basic Materials

• Laptop with internet access.
• Spreadsheet software such as Google Sheets or Excel.
• R or Python for data cleaning and plots.
• Access to BLS CPS public microdata through IPUMS CPS or BLS files.
• A state minimum-wage timeline from the U.S. Department of Labor or state labor sites.
• A notebook for tracking policy dates, sample rules, and code decisions.

Advanced Materials

• University computer account with R, Python, or both.
• Access to a statistical package with synthetic-control tools.
• IPUMS CPS or another secure CPS microdata workspace.
• Detailed state policy file with minimum-wage effective dates and exemptions.
• High-memory workstation for repeated placebo runs.
• Reference copy of the CPS codebook and survey weights documentation.

Software & Tools

• R: Fits synthetic-control models, runs placebo tests, and makes trend plots.
• Python: Cleans CPS records, merges policy data, and automates sensitivity checks.
• RStudio: Organizes scripts, outputs, and data notes in one workspace.
• IPUMS CPS extract system: Builds a harmonized CPS file with the variables you need.
• Google Sheets: Helps track policy dates, state groups, and summary statistics.

Experiment Steps

1. Define the exact teen age band, state, and policy window you will study.
2. Build a comparison set of states that match your treated state before the wage change.
3. Prepare the CPS microdata so the outcome, weights, and sample rules are aligned.
4. Fit the synthetic control and check whether it tracks the treated state before the policy date.
5. Run placebo tests, alternative donor pools, and subgroup splits to probe whether the pattern holds.
6. Turn the estimates into one clear policy story with a graph, a table, and a plain-language claim.

Common Pitfalls

• Comparing raw teen job counts instead of participation rates, which hides population changes across states.
• Ignoring the policy effective date, which mixes pre-change and post-change months in the same estimate.
• Choosing donor states with very different pre-trends, which makes the synthetic control a poor stand-in.
• Forgetting to apply CPS survey weights, which can bias statewide estimates.
• Treating one state change as proof of causation without placebo tests, which leaves the result weak.

What Makes This Competitive

A strong version goes beyond asking whether teen work changed, it asks where, for whom, and under what policy setting the shift happened. Use placebo states, multiple donor pools, and pre-trend checks so your estimate does not depend on one lucky comparison. Add subgroup analysis by age or school status, and you turn a simple policy chart into a deeper causal story. Judges tend to like projects that show clean identification and careful sensitivity checks.

Project Variations

• Compare teen labor-force participation after minimum-wage hikes in border counties versus the rest of the state.
• Replace labor-force participation with teen hours worked to test whether the policy changes job holding instead of job search.
• Split the sample by school status to see whether enrolled teens and out-of-school teens respond differently.

Learn More

• BLS Current Population Survey Handbook: Search BLS for labor-force definitions, survey design, and weighting notes.
• IPUMS CPS: Search IPUMS for harmonized CPS microdata, variable descriptions, and download tools.
• U.S. Department of Labor minimum wage history: Search DOL and state labor sites for policy dates and changes.
• MIT OpenCourseWare Econometrics: Search MIT OpenCourseWare for regression, panel data, and causal inference notes.
• American Economic Review: Search Google Scholar or a library database for minimum-wage and teen employment studies.
• Journal of Human Resources: Search Google Scholar or a library database for labor-market research on teens and policy changes.